Dr Rosenberg received his medical degree from the University of California, Irvine, and is currently practicing as a third-year fellow in the Division of Hematology/Oncology at Scripps Clinic.
Dr Sigal received his medical degree from the University of California, Los Angeles, and is currently practicing in the Division of Hematology/Oncology at Scripps Clinic Medical Group. He is Principal Investigator of the Pancreas and Biliary Cancer Program at Scripps Clinic.
In 2012, pancreatic ductal adenocarcinoma afflicted an estimated 43,920 people in the United States, and approximately 37,390 people died of this disease. Despite being only the tenth most common cancer, it is the fourth most common cause of cancer death.1 About 20% of patients will present with localized, resectable pancreatic cancer. Although adjuvant chemotherapy with gemcitabine or 5-fluorouracil (5-FU) confers a small survival advantage, the vast majority of these patients will relapse and die of their disease. The other 80% of patients will have either de novo metastasis or locally advanced cancer making potentially curative resection impossible.2 Overall, only about 4% of patients will be alive and disease free at 5 years.3-5 It is clear that pancreatic cancer represents an urgent unmet medical need for more effective therapies.
Two important developments that may finally alter the natural history of incurable pancreatic cancer have been reported over the past couple of years. In a phase 3 trial, patients randomized to FOLFIRINOX (combination bolus and infusional 5-FU, irinotecan, and oxaliplatin) had significant improvements in median overall survival (OS) (11.1 vs 6.8 months; P<.001) and overall response rates (32% vs 9%; P<.001) compared with single-agent gemcitabine, respectively.6 Recently presented at the 2013 Gastrointestinal Cancers Symposium was a large phase 3 trial (MPACT) that randomized patients to combination nab-paclitaxel and gemcitabine versus single-agent gemcitabine. Patients in the combination therapy cohort had improved median OS (8.5 vs 6.7 months; P=.000015) and overall response rates (23% vs 7%, P=1.1×10–10).7 These are clinically meaningful improvements over published reports of single-agent gemcitabine or the combination of gemcitabine and erlotinib.8,9
With the advent of potentially multiple effective therapeutic regimens, incorporation of pancreatic cancer predictive biomarkers into clinical care may enable the selection of the most effective therapy for a specific tumor profile. Predictive, as opposed to simply prognostic, biomarkers are key for achieving this goal. A prognostic biomarker predicts outcome differences independent of any intervention, whereas a predictive biomarker predicts outcome differences based on a specific intervention, making the latter central for determining therapy.
Some of the more promising biomarkers include human equilibrative nucleoside transporter 1 (hENT1), secreted protein acidic and rich in cysteine (SPARC), deoxycytidine kinase (dCK), and ribonucleotide reductase subunit M1 (RRM1) (Figure). Each of these proteins may be integral for chemotherapeutic agent uptake, metabolism, or cytotoxicity, making them putative predictive biomarkers. Solid evidence supporting a biomarker’s predictive value usually follows a pathway beginning with preclinical studies (in vitro, ex vivo, animal models) and clinical prognostic data before predictive data can be generated. These 4 biomarkers are well along this path, and we feel that an overall review of them is now appropriate.
hENT1
Gemcitabine is a hydrophilic pyrimidine nucleoside analogue that requires specialized membrane nucleoside transporter proteins to efficiently cross the plasma membrane and enter cells.10 hENT1 is found in most cell types and mediates concentration-dependent gemcitabine cellular uptake.11,12 Gemcitabine levels are increased in pancreatic cancer cell lines incubated with gemcitabine that display high hENT1 expression.13 In vitro gemcitabine toxicity is determined by hENT1 expression and activity patterns. High hENT1-expressing cell lines incubated with dipyridamole, a potent inhibitor of hENT1 activity, were resistant to gemcitabine due to its reduced cellular uptake.14,15 Conversely, in a pancreatic tumor xenograft mouse model that used 5-FU to upregulate hENT1, a significant cellular growth inhibition was noted when gemcitabine was administered subsequently.16 Interestingly, a phase 3 trial did not show improved OS among patients randomized to combination capecitabine, an oral 5-FU prodrug, and gemcitabine compared with gemcitabine alone, although outcomes according to hENT1 status were not reported.17 As will be seen with all 4 biomarkers evaluated in this article, none of the clinical studies evaluating the prognostic or predictive value of hENT1 utilized consistent hENT1 assays or cutoff values. Although similar antibodies were employed for immunohistochemistry (IHC), different scoring systems were utilized – one trial used a 0 to 2 scale based on adjacent islet of Langerhans cell and lymphocyte staining, while others used a 0 to 3 scale based on the percentage of cells at each staining level. Another study used reverse transcription-polymerase chain reaction (RT-PCR) to determine high and low values according to a ratio with the glyceraldehyde-3-phosphate dehydrogenase reference gene.18-25
Several retrospective analyses have correlated pancreatic cancer patient outcomes to hENT1 expression. However, because all patients in these studies received gemcitabine and no untreated control groups existed, they can only support the role of hENT1 as a prognostic marker. RT-PCR analysis for hENT1 expression was performed on 102 microdissected pancreatic cancer samples from patients with localized to metastatic disease, and all patients were treated with gemcitabine.18 Patients with high hENT1 levels had a significantly longer median OS versus those in the lower expression subgroup at 25.7 versus 8.5 months (P<.001).18 Additionally, patients with high versus low hENT1 expression had a statistically significant advantage when comparing disease-free survival (DFS) (20.4 vs 9.3 months) in the adjuvant setting and time to progression (12.7 vs 5.9 months; P=.02) in the metastatic setting.18 Multivariate analysis, comprising adjuvant or palliative treatment, tumor grade, cytidine deaminase expression, and hENT1 expression, confirmed the prognostic significance of hENT1 expression as an independent factor associated with survival.18 Two additional retrospective studies measured hENT1 by IHC in the metastatic and adjuvant setting and confirmed the prognostic power of hENT1.19,20 However, a more recent study that randomly selected 95 specimens from patients who had undergone pancreaticoduodenectomy was unable to correlate hENT1 expression to survival.21 Cross-study comparisons are always difficult, especially for small retrospective trials, but this trial does highlight the importance of developing optimal hENT1 detection assays, defining threshold hENT1 expression criteria, and contemplating the possibility that its prognostic value may differ in the adjuvant or metastatic setting.
Two retrospective studies raised the possibility of hENT1 as a predictive biomarker. Farrell and colleagues22 evaluated hENT1 expression by IHC from 229 of 538 patients who had undergone surgical resection. These patients were enrolled in RTOG 97-04,23 a phase 3 adjuvant study evaluating the impact of adding gemcitabine to adjuvant 5-FU chemoradiation on survival. High hENT1 expression was associated with both improved OS and DFS (P=.01 and P=.04, respectively) only in patients who received gemcitabine.22 Another study analyzed hENT1 by IHC in 434 specimens from patients who had also undergone curative surgery.24
Patients with high hENT1 expression who also received adjuvant gemcitabine had improved OS (P<.001) compared with those who did not receive gemcitabine. Patients with low hENT1 did not benefit from gemcitabine.17 The only prospective study (LEAP) was a phase 2 trial that randomized 360 patients to gemcitabine or CO-101, a gemcitabine-lipid conjugate that is able to enter cells independent of hENT1.25 Inclusion criteria required patients to have metastatic pancreatic cancer and possess tumors with low hENT1 expression. A recent company media release indicated that no difference in survival was found between patient cohorts.26 These negative findings call into question whether drug delivery is the mechanism of action for the putative predictive value of low hENT1 status for gemcitabine therapy and reinforce the need for inclusion of both biomarker-positive and biomarker-negative cohorts in prospective, randomized trials to better understand the impact of the markers in question.
SPARC
Secreted protein acidic and rich in cysteine (SPARC), also referred to as osteonectin, is a calcium-binding glycoprotein involved in cell-matrix interactions. SPARC is involved in many biological processes including cellular differentiation, tissue remodeling, cell migration, cell invasion, morphogenesis, and angiogenesis.27-29 SPARC-null mice have reduced pancreatic tumor cell apoptosis and enhanced tumor progression, demonstrating a growth inhibitory function.30,31 SPARC is generally expressed in normal pancreatic tissue. However, aberrant promoter hypermethylation is likely responsible for loss of SPARC expression within pancreatic cancer. SPARC overexpression is actually induced in the peritumoral stromal fibroblasts and extracellular matrix immediately adjacent to the cancer.32-34 The clinical importance of SPARC in pancreatic cancer was first suggested in a retrospective study of 299 pancreaticoduodenectomy specimens that were stained for SPARC by IHC. Patients whose pancreatic cancer peritumoral fibroblasts expressed SPARC had a significantly worse median OS than those whose peritumoral stroma did not (15 vs 30 months; P<.001).35
An albumin-bound nanoparticle form of paclitaxel, nab-paclitaxel, binds SPARC via albumin, potentially directly delivering paclitaxel to tumors with increased SPARC expression.36 SPARC has therefore been postulated to be a potential predictive biomarker for nab-paclitaxel. The phase 1/2 trial that determined the maximum tolerated dose of combination nab-paclitaxel and gemcitabine included a preplanned subgroup analysis of SPARC expression by IHC scored by an exhaustive methodology.37 Patients with high SPARC expression had a significantly prolonged median OS compared with those with low SPARC expression (17.8 vs 8.1 months; P=.0431).37 In contrast to the previously described retrospective study showing that peritumoral SPARC expression correlated with worse outcome, this trial showed that patients with increased peritumoral SPARC actually had improved survival. This discrepancy may result from inconsistencies in detecting SPARC expression. Alternatively, this phase 1/2 trial was meticulous in determining SPARC expression, suggesting that nab-paclitaxel may more effectively deliver gemcitabine to the tumor, overcoming the deleterious impact of SPARC overexpression, and reinforcing the possibility that SPARC is a predictive biomarker for benefit from the combination of nab-paclitaxel and gemcitabine. A randomized phase 3 trial of the combination of nab-paclitaxel and gemcitabine was recently reported at the 2013 Gastrointestinal Cancers Symposium, but SPARC analysis was not included.7 A more definitive view of the predictive value of SPARC will emerge when the data are released.
dCK
Deoxycytidine kinase (dCK) is an intracellular enzyme that serves as the rate-limiting step in the phosphorylation of gemcitabine to its active metabolites, gemcitabine monophosphate, diphosphate, and triphosphate.38 Overexpression of dCK has demonstrated increased gemcitabine sensitivity in various solid tumor cell lines, including breast, non–small cell lung, colon, ovarian, head and neck, and pancreatic cancers.39,40 Human colon carcinoma xenografts transfected with the dCK gene resulted in increased gemcitabine triphosphate cellular accumulation, prolonged elimination kinetics, and potentiated an in vivo tumor response to gemcitabine.41 Accordingly, dCK downregulation induced gemcitabine resistance in pancreatic cancer cells.42 Among the clinical studies evaluating the prognostic or predictive value of dCK, no consistency existed for either dCK assays or cutoff values. Even when a similar antibody was used for IHC, different scoring systems were employed – one trial used a 0 to 3 scale based on adjacent lymphocyte staining, and another used an intensity score defined as the product of a 0 to 2 scale and the percentage of cells at each staining level. Studies that used RT-PCR either correlated it to a ?-actin control or determined high and low values according to the average RT-PCR value for the samples being evaluated.43-46
There is some evidence supporting the role of dCK as a prognostic biomarker. A retrospective review was performed on 44 specimens stained for dCK by IHC from patients treated with gemcitabine in the adjuvant or metastatic setting. Patients with high dCK expression had a significant improvement in median OS (22 vs 15 months; P<.009) compared with those with low dCK expression. High dCK expressers maintained their survival advantage even after adjusting for the progression-free survival (PFS) benefit after gemcitabine therapy (12 vs 10 months; P<.04).43 Prior to initiation of palliative gemcitabine in 35 patients with locally advanced pancreatic cancer, biopsies were obtained by endoscopic ultrasound fine-needle aspiration, and dCK mRNA was determined by RT-PCR. High dCK mRNA levels correlated to gemcitabine efficacy, while low dCK mRNA levels did not.44 Forty-five patients with localized pancreatic cancer who underwent curative resection as part of 2 separate phase 2 trials had tumor specimens stained for dCK by IHC. Median OS was 13.2 months for patients with low dCK expression and was not reached for those with high dCK expression (P=.0008). DFS was 6.3 months for low dCK expressers and 46.8 months for high dCK expressers (P=.0003). dCK expression was the only significant independent marker for OS and DFS in comparison with lymph node involvement, lymph node ratio, and greatest tumor
diameter.45
Two retrospective studies report on dCK as a predictive biomarker. The expression of dCK by IHC was determined among 434 pancreatic cancer patients who underwent curative resection, with 243 of these patients administered adjuvant gemcitabine. Patients with high dCK who received adjuvant gemcitabine had improved survival compared with those who did not (hazard ratio 0.57; 95% CI, 0.41-0.78; P=.001). Patients with low dCK did not derive any benefit from gemcitabine (P=.66).24 RT-PCR was used to determine dCK mRNA levels in 40 patients who received adjuvant gemcitabine and 30 who did not after curative resection. High dCK mRNA expression was associated with a significantly longer DFS (P=.0067) in the gemcitabine-treated group.46
RRM1
Ribonucleotide reductase subunit M1 (RRM1) is the regulatory subunit of ribonucleotide reductase, an enzyme essential for cell survival that catalyzes the rate-limiting step of the DNA biosynthetic pathway, converting ribonucleoside diphosphate to deoxyribonucleoside diphosphate.47 RRM1 is the intracellular target of gemcitabine 5?-diphosphate, a phosphorylated metabolite of gemcitabine.48 RRM1 inhibition reduces the cellular concentration of deoxynucleoside triphosphates, blocking DNA synthesis.49 Gemcitabine resistance in non–small cell lung cancer and colon cancer cell lines was associated with increased RRM1 gene expression.50,51 Genetically modified lung cancer cell lines with increasing RRM1 expression displayed reduced gemcitabine efficacy.52 Finally, suppression of RRM1 expression by RNA interference restored gemcitabine sensitivity in a pancreatic cancer cell line.24 Increased RRM1 expression likely overcomes gemcitabine activity by increasing the deoxyribonucleotide triphosphate pool, competitively inhibiting the incorporation of gemcitabine triphosphate into DNA.53 RRM1 can also serve as a “molecular sink” by removing intracellular gemcitabine diphosphate, reducing the accumulation of active gemcitabine triphosphate.50 Again, the clinical studies evaluating the prognostic or predictive value of RRM1 had varying assays and cutoff values. Though similar antibodies were used for IHC, one study used a 0 to 3 scoring system based on the percentage of cytoplasmic staining, while the other was determined by automated quantitative analysis, a fluorescence-based IHC method. Studies that used RT-PCR correlated RRM1 mRNA to the reference genes ?-actin or glyceraldehyde 3-phosphate dehydrogenase, or determined high and low values according to the average RT-PCR value for the samples being evaluated.44,54-56
Figure
Several reports support the prognostic value of RRM1 in pancreatic cancer. In a retrospective analysis, 68 pancreatic cancer patients underwent curative R0 resections with only 5 receiving adjuvant gemcitabine. Patients with low RRM1 expression determined by IHC had significantly decreased 3-year OS compared with those with high expression (46% vs 28%; P=.0196).
Multivariate analysis noted that RRM1 expression was the only independent determinant of OS. Among the 23 patients treated with gemcitabine at disease recurrence, only patients with low RRM1 expression had a significant survival benefit (P=.0010), suggesting that low RRM1 is predictive for benefit from gemcitabine. However, this was still unclear since all patients were treated with gemcitabine and an untreated control group did not exist.54 Another retrospective report reinforced these results. Eighteen pancreatic cancer patients who had curative resection and did not receive adjuvant gemcitabine were administered gemcitabine at the time of their recurrence. RRM1 mRNA levels were determined in the primary tumor specimens, and patients with low levels had significantly improved OS (P=.016).55
Table
However, these findings were not confirmed in 2 other studies. In a retrospective analysis of RRM1 expression by both IHC and RT-PCR in 94 patients with resected pancreatic cancer, RRM1 expression by either technique was not prognostic for OS, and low levels were not predictive for benefit from gemcitabine treatment.56 Finally, a small trial determined RRM1 mRNA expression by endoscopic ultrasound fine-needle aspiration in patients with locally advanced pancreatic cancer prior to the initiation of gemcitabine chemotherapy. RRM1 expression levels were not correlated with benefit from gemcitabine.44
Conclusion
After many years of fruitless clinical investigations, important progress is finally being made in the treatment of pancreatic cancer. For patients with an adequate performance status, single-agent gemcitabine is yielding to combination therapies, such as FOLFIRINOX, and more recently combination gemcitabine and nab-paclitaxel. For the first time, oncologists who treat patients with pancreatic cancer will be faced with the dilemma of drug sequencing. Unique mechanisms of action from these regimens, and future regimens, may be exploited to confer differential clinical benefits to specific patient cohorts.
We have reviewed a selection of promising pancreatic cancer biomarkers that may eventually help guide clinical decision making and drug sequencing (Table). hENT1, RRM1, and dCK directly impact gemcitabine metabolism. SPARC is expressed in the pancreatic cancer peritumoral stroma and is targeted by albumin, potentially enhancing the activity of combination nab-paclitaxel and gemcitabine. Unfortunately, these potentially predictive pancreatic cancer biomarkers are not yet ready to be incorporated into patient care. Even for hENT1, the biomarker with the most clinically advanced data, most of the predictive data are from small retrospective studies with mixed adjuvant and metastatic populations. In the only prospective phase 3 trial, hENT1 expression was not predictive for benefit from a novel gemcitabine-lipid conjugate (CO-101), potentially undermining its predictive role for nongemcitabine therapies.26 SPARC is the only other biomarker with prospective data in a small phase 1/2 study that suggested elevated SPARC expression was predictive for benefit from the combination of nab-paclitaxel and gemcitabine.37 However, SPARC data were not included in the recent presentation of the phase 3 trial for this regimen, preventing any definitive conclusions of SPARC predictivity.7
The lack of large prospective trials validating these biomarkers should not necessarily impede their use. KRAS, the most important predictive colorectal cancer predictive biomarker, gained acceptance with only retrospective trials.57,58 However, these retrospective analyses should evolve from prospectively randomized trials of sufficient size and adequate control groups. A further difficulty in evaluating pancreatic cancer biomarkers, whether retrospectively or prospectively, is the general lack of consensus for accurate biomarker assays and inconsistent threshold values to determine positivity or negativity. Exacerbating this challenge is the limited amount of cancer tissue available from endoscopic ultrasound biopsies of the pancreatic tumor. This is particularly true in the locally advanced or neoadjuvant setting where large metastatic deposits are not available for core biopsies and the only tissue available is from an endoscopically attained fine-needle aspiration. However, the data that exist for these biomarkers are certainly hypothesis generating for their predictive role in pancreatic cancer. Future large retrospective analyses should ensure that inconsistencies in assays, threshold levels, and patient population and appropriately matched control groups are addressed. Also, randomized trials of new pancreatic cancer–targeted agents that prospectively incorporate reliable biomarker assays are the most promising approach for finally and definitively bringing predictive biomarkers into patient care.
References
1. Siegel R, Naishadham D, Jemal A. Cancer statistics, 2012. CA Cancer J Clin. 2012;62:10-29.
2. Vincent A, Herman J, Schulick R, et al. Pancreatic cancer. Lancet. 2011;378:607-620.
3. Yeo CJ, Cameron JL, Lillemoe KD, et al. Pancreaticoduodenectomy for cancer of the head of the pancreas. 201 patients. Ann Surg. 1995;221:721-731.
4. Geer RJ, Brennan MF. Prognostic indicators for survival after resection of pancreatic adenocarcinoma. Am J Surg. 1993;165:68-72.
5. Neoptolemos JP, Stocken DD, Friess H, et al. A randomized trial of chemoradiotherapy and chemotherapy after resection of pancreatic cancer. N Engl J Med. 2004;350:1200-1210.
6. Conroy T, Desseigne F, Ychou M, et al. FOLFIRINOX versus gemcitabine for metastatic pancreatic cancer. N Engl J Med. 2011;364:1817-1825.
7. Von Hoff DD, Ervin TJ, Arena FP, et al. Randomized phase III study of weekly nab-paclitaxel plus gemcitabine versus gemcitabine alone in patients with metastatic adenocarcinoma of the pancreas (MPACT). J Clin Oncol. 2012;30(suppl 34). Abstract LBA 148.
8. Burris HA 3rd, Moore MJ, Andersen J, et al. Improvements in survival and clinical benefit with gemcitabine as first-line therapy for patients with advanced pancreas cancer: a randomized trial. J Clin Oncol. 1997;15:2403-2413.
9. Moore MJ, Goldstein D, Hamm J, et al. Erlotinib plus gemcitabine compared with gemcitabine alone in patients with advanced pancreatic cancer: a phase III trial of the National Cancer Institute of Canada Clinical Trials Group. J Clin Oncol. 2007;25:1960-1966.
10. Mackey JR, Baldwin SA, Young JD, et al. Nucleoside transport and its significance for anticancer drug resistance. Drug Resist Updat. 1998;1:310-324.
11. Pennycooke M, Chaudary N, Shuralyova I, et al. Differential expression of human nucleoside transporters in normal and tumor tissue. Biochem Biophys Res Commun. 2001;280:951-959.
12. Young JD, Yao SY, Sun L, et al. Human equilibrative nucleoside transporter (ENT) family of nucleoside and nucleobase transporter proteins. Xenobiotica. 2008;38:995-1021.
13. Garcia-Manteiga J, Molina-Arcas M, Casado FJ, et al. Nucleoside transporter profiles in human pancreatic cancer cells: role of hCNT1 in 2?,2?-difluorodeoxycytidine-induced cytotoxicity. Clin Cancer Res. 2003;9:5000-5008.
14. Mackey JR, Mani RS, Selner M, et al. Functional nucleoside transporters are required for gemcitabine influx and manifestation of toxicity in cancer cell lines. Cancer Res. 1998;58:4349-4357.
15. Mori R, Ishikawa T, Ichikawa Y, et al. Human equilibrative nucleoside transporter 1 is associated with the chemosensitivity of gemcitabine in human pancreatic adenocarcinoma and biliary tract carcinoma cells. Oncol Rep. 2007;17:1201-1205.
16. Nakahira S, Nakamori S, Tsujie M, et al. Pretreatment with S-1, an oral derivative of 5-fluorouracil, enhances gemcitabine effects in pancreatic cancer xenografts. Anticancer Res. 2008;28:179-186.
17. Cunningham D, Chau I, Stocken DD, et al. Phase III randomized comparison of gemcitabine versus gemcitabine plus capecitabine in patients with advanced pancreatic cancer. J Clin Oncol. 2009;27:5513-5518.
18. Giovannetti E, Del Tacca M, Mey V, et al. Transcription analysis of human equilibrative nucleoside transporter-1 predicts survival in pancreas cancer patients treated with gemcitabine. Cancer Res. 2006;66:3928-3935.
19. Spratlin J, Sangha R, Glubrecht D, et al. The absence of human equilibrative nucleoside transporter 1 is associated with reduced survival in patients with gemcitabine-treated pancreas adenocarcinoma. Clin Cancer Res. 2004;10:6956-6961.
20. Maréchal R, Mackey JR, Lai R, et al. Human equilibrative nucleoside transporter 1 and human concentrative nucleoside transporter 3 predict survival after adjuvant gemcitabine therapy in resected pancreatic adenocarcinoma. Clin Cancer Res. 2009;15:2913-2919.
21. Fisher SB, Patel SH, Bagci P, et al. An analysis of ERCC1, hENT1, RRM1, and RRM2 expression in resected pancreas adenocarcinoma: implications for adjuvant treatment. J Clin Oncol. 2012;30(suppl 4). Abstract 206.
22. Farrell JJ, Elsaleh H, Garcia M, et al. Human equilibrative nucleoside transporter 1 levels predict response to gemcitabine in patients with pancreatic cancer. Gastroenterology. 2009;136:187-195.
23. Regine WF, Winter KA, Abrams RA, et al. Fluorouracil vs gemcitabine chemotherapy before and after fluorouracil-based chemoradiation following resection of pancreatic adenocarcinoma: a randomized controlled trial. JAMA. 2008;299:1019-1026.
24. Maréchal R, Bachet J, Mackey JR, et al. Prediction of gemcitabine benefit after curative-intent resection of pancreatic adenocarcinoma using HENT1 and dCK protein expression. J Clin Oncol. 2011;29(suppl). Abstract 4024.
25. Business Wire. Clovis Oncology completes enrollment in pivotal LEAP trial of CO-101 versus gemcitabine in metastatic pancreatic cancer. www.businesswire.com/news/home/20120326005268/en/Clovis-Oncolo gy-Completes-Enrollment-Pivotal-LEAP-Trial. Accessed April 28, 2013.
26. Business Wire. Clovis Oncology announces negative outcome of CO-101 in pivotal LEAP pancreatic cancer study. www.businesswire.com/news/home/20121111005032/en/Clovis-Oncology-Announces-Negative- Outcome-CO-101-Pivotal. Accessed April 28, 2013.
27. Bradshaw AD, Sage EH. SPARC, a matricellular protein that functions in cellular differentiation and tissue response to injury. J Clin Invest. 2001;107:1049-1054.
28. Jacob K, Webber M, Benayahu D, et al. Osteonectin promotes prostate cancer cell migration and invasion: a possible mechanism for metastasis to bone. Cancer Res. 1999;59:4453-4457.
29. Brekken RA, Sage EH. SPARC, a matricellular protein: at the crossroads of cell-matrix communication. Matrix Biol. 2001;19:816-827.
30. Puolakkainen PA, Brekken RA, Muneer S, et al. Enhanced growth of pancreatic tumors in SPARC-null mice is associated with decreased deposition of extracellular matrix and reduced tumor cell apoptosis. Mol Cancer Res. 2004;2:215-224.
31. Brekken RA, Puolakkainen P, Graves DC, et al. Enhanced growth of tumors in SPARC null mice is associated with changes in the ECM. J Clin Invest. 2003;111:487-495.
32. Ueki T, Toyota M, Sohn T, et al. Hypermethylation of multiple genes in pancreatic adenocarcinoma. Cancer Res. 2000;60:1835-1839.
33. Sato N, Fukushima N, Maehara N, et al. SPARC/osteonectin is a frequent target for aberrant methylation in pancreatic adenocarcinoma and a mediator of tumor-stromal interactions. Oncogene. 2003;22:5021-5030.
34. Guweidhi A, Kleeff J, Adwan H, et al. Osteonectin influences growth and invasion of pancreatic cancer cells. Ann Surg. 2005;242:224-234.
35. Infante JR, Matsubayashi H, Sato N, et al. Peritumoral fibroblast SPARC expression and patient outcome with resectable pancreatic adenocarcinoma. J Clin Oncol. 2007;25:319-325.
36. Von Hoff DD, Ramanathan R, Borad M, et al. SPARC correlation with response to gemcitabine (G) plus nab-paclitaxel (nab-P) in patients with advanced metastatic pancreatic cancer: a phase I/II study. J Clin Oncol. 2009;27(suppl). Abstract 4525.
37. Von Hoff DD, Ramanathan RK, Borad MJ, et al. Gemcitabine plus nab-paclitaxel is an active regimen in patients with advanced pancreatic cancer: a phase I/II trial. J Clin Oncol. 2011;29:4548-4554.
38. Mini E, Nobili S, Caciagli B, et al. Cellular pharmacology of gemcitabine. Ann Oncol. 2006;17(suppl 5):v7-v12.
39. Hapke DM, Stegmann AP, Mitchell BS. Retroviral transfer of deoxycytidine kinase into tumor cell lines enhances nucleoside toxicity. Cancer Res. 1996;56:2343-2347.
40. Kroep JR, Loves WJ, van der Wilt CL, et al. Pretreatment deoxycytidine kinase levels predict in
vivo gemcitabine sensitivity. Mol Cancer Ther. 2002;1:371-376.
41. Blackstock AW, Lightfoot H, Case LD, et al. Tumor uptake and elimination of 2?,2?-difluoro-2?-deoxycytidine (gemcitabine) after deoxycytidine kinase gene transfer: correlation with in vivo tumor response. Clin Cancer Res. 2001;7:3263-3268.
42. Ohhashi S, Ohuchida K, Mizumoto K, et al. Down-regulation of deoxycytidine kinase enhances acquired resistance to gemcitabine in pancreatic cancer. Anticancer Res. 2008;28:2205-2212.
43. Sebastiani V, Ricci F, Rubio-Viqueira B, et al. Immunohistochemical and genetic evaluation of deoxycytidine kinase in pancreatic cancer: relationship to molecular mechanisms of gemcitabine resistance and survival. Clin Cancer Res. 2006;12:2492-2497.
44. Ashida R, Nakata B, Shigekawa M, et al. Gemcitabine sensitivity-related mRNA expression in endoscopic ultrasound-guided fine-needle aspiration biopsy of unresectable pancreatic cancer. J Exp Clin Cancer Res. 2009;28:83.
45. Maréchal R, Mackey JR, Lai R, et al. Deoxycitidine kinase is associated with prolonged survival after adjuvant gemcitabine for resected pancreatic adenocarcinoma. Cancer. 2010;116:5200-5206.
46. Fujita H, Ohuchida K, Mizumoto K, et al. Gene expression levels as predictive markers of outcome in pancreatic cancer after gemcitabine-based adjuvant chemotherapy. Neoplasia. 2010;12:807-817.
47. Reichard P. From RNA to DNA, why so many ribonucleotide reductases? Science. 1993;260:1773-1777.
48. van der Donk WA, Yu G, Pérez L, et al. Detection of a new substrate-derived radical during inactivation of ribonucleotide reductase from Escherichia coli by gemcitabine 5?-diphosphate. Biochemistry. 1998;37:6419-6426.
49. Heinemann V, Xu YZ, Chubb S, et al. Inhibition of ribonucleotide reduction in CCRF-CEM cells by 2?,2?-difluorodeoxycytidine. Mol Pharmacol. 1990;38:567-572.
50. Davidson JD, Ma L, Flagella M, et al. An increase in the expression of ribonucleotide reductase large subunit 1 is associated with gemcitabine resistance in non-small cell lung cancer cell lines. Cancer Res. 2004;64:3761-3766.
51. Bergman AM, Eijk PP, van Haperen VW, et al. In vivo induction of resistance to gemcitabine results in increased expression of ribonucleotide reductase subunit M1 as a major determinant. Cancer Res. 2005;65:9510-9516.
52. Bepler G, Kusmartseva I, Sharma S, et al. RRM1-modulated in vitro and in vivo efficacy of gemcitabine and platinum in non-small cell lung cancer. J Clin Oncol. 2006;24:4731-4737.
53. Plunkett W, Huang P, Searcy CE, et al. Gemcitabine: preclinical pharmacology and mechanisms of action. Semin Oncol. 1996;23(suppl 10):3-15.
54. Akita H, Zheng Z, Takeda Y, et al. Significance of RRM1 and ERCC1 expression in resectable pancreatic adenocarcinoma. Oncogene. 2009;28:2903-2909.
55. Nakahira S, Nakamori S, Tsujie M, et al. Involvement of ribonucleotide reductase M1 subunit overexpression in gemcitabine resistance of human pancreatic cancer. Int J Cancer. 2007;120:1355-1363.
56. Valsecchi ME, Holdbrook T, Leiby BE, et al. Is there a role for the quantification of RRM1 and ERCC1 expression in pancreatic ductal adenocarcinoma? BMC Cancer. 2012;12:104.
57. Karapetis CS, Khambata-Ford S, Jonker DJ, et al. K-ras mutations and benefit from cetuximab in advanced colorectal cancer. N Engl J Med. 2008;359:1757-1765.
58. Amado RG, Wolf M, Peeters M, et al. Wild-type KRAS is required for panitumumab efficacy in patients with metastatic colorectal cancer. J Clin Oncol. 2008;26:1626-1634.